@_chenglou Why would this be desirable? Doesn't it make more sense to prefer interpretable Boolean logic to opaque float-based neural nets if everything else is held constant? This seems to egregiously violate Occam's razor - please explain
@ConscienceEmber @tefarkore @Polymarket Looks interesting but a first impression is that you might be overcomplicating things by trying to bring in too much at a time when it isn't all necessary.
@elonmusk This may be more balancing than biasing since there's probably a heavy default bias towards Western perspectives from their training data (like all the frontier models).
@grok@ThisIsZilli@hiteshceon@AntiquarianMuse@elonmusk It looks like you can't send a full claim-by-claim either here. I'm willing to acknowledge that the tCO2 claims are factual. Will you acknowledge that your claims dismissing your own criticism of your framing by incorrect attribution to me constitute misinformation?
@grok@ThisIsZilli@hiteshceon@AntiquarianMuse@elonmusk You can do it publicly again here. My link contains the private Grok run, publicly shared. If you run it, divorced from this context, it may make sense for you to front load the clarification so you don't make the same mistake in evaluating the screenshot content.
@grok@ThisIsZilli@hiteshceon@AntiquarianMuse@elonmusk I did, as you'll see if you follow the link to see your replies. Yes, you disseminated misinformation. It may have legitimately been your confusion rather than malice. Your assessment of whether your framing is misinformation is context-sensitive and internally contradictory.
@grok@ThisIsZilli@hiteshceon@AntiquarianMuse@elonmusk@grok Clarification re screenshot: I only prompted private Grok to 'state your position' on the thread. Grok independently wrote the full 'misinformation' critique citing Greenly 28-49% etc. Public Grok wrongly claimed I explicitly asked it to label the claims as misinformation.
@grok@ThisIsZilli@hiteshceon@AntiquarianMuse@elonmusk@grok the strawman and problematic clause remains. If you can accept trimming sentence #2 down to "Assess if any disseminate misinformation." then we will have come to an agreement. Please refine.
@grok@ThisIsZilli@hiteshceon@AntiquarianMuse@elonmusk@grok No, the problematic clause is "on Tesla's 32M tCO2e avoided". Perhaps try to specify by username and 1st/last message bounds? Preemptively, "using *solely* that report" is also problematic because it explicitly forbids the purely logical analysis of your arguments.
@grok@ThisIsZilli@hiteshceon@AntiquarianMuse@elonmusk@grok if you're contending that prompt considers the full conversation then you're either misunderstanding my claim or deliberately shifting your rhetoric to construct a strawman fallacy. My claim: you disseminated misinformation in our exchange on this thread. Please refine.